Dockerfile

# A2C Knowledge Base Docker Image
#
# Supports: PDF, DOCX, PPTX, HTML, Markdown, Images, Audio (Whisper ASR), Video (ffmpeg)
#
# Build:
#   docker build -t a2c-knowledge .
#
# Run:
#   docker run -it --rm -v $(pwd)/data:/app/data a2c-knowledge
#
# With GPU (NVIDIA):
#   docker run -it --rm --gpus all -v $(pwd)/data:/app/data a2c-knowledge

FROM python:3.11-slim

# Set environment variables
ENV PYTHONUNBUFFERED=1 \
    PYTHONDONTWRITEBYTECODE=1 \
    PIP_NO_CACHE_DIR=1 \
    PIP_DISABLE_PIP_VERSION_CHECK=1

# Install system dependencies
RUN apt-get update && apt-get install -y --no-install-recommends \
    # Build tools
    build-essential \
    # ffmpeg for video/audio processing
    ffmpeg \
    # PDF processing dependencies
    poppler-utils \
    tesseract-ocr \
    tesseract-ocr-chi-sim \
    tesseract-ocr-chi-tra \
    # Image processing
    libgl1-mesa-glx \
    libglib2.0-0 \
    # Cleanup
    && rm -rf /var/lib/apt/lists/*

# Set working directory
WORKDIR /app

# Install uv for faster package installation
RUN pip install uv

# Copy project files
COPY pyproject.toml .
COPY knowledge/ knowledge/
COPY config/ config/
COPY generator/ generator/
COPY models/ models/
COPY renderer/ renderer/
COPY utils/ utils/
COPY exporter/ exporter/

# Install Python dependencies
RUN uv pip install --system -e ".[asr]"

# Pre-download models (optional, makes first run faster)
# Uncomment to include models in the image (increases image size significantly)
# RUN python -c "from sentence_transformers import SentenceTransformer; SentenceTransformer('Qwen/Qwen3-Embedding-0.6B')"

# Create data directory for ChromaDB
RUN mkdir -p /app/data/chroma

# Default command
CMD ["python", "-c", "from knowledge import KnowledgePipeline; print('A2C Knowledge Base ready!')"]